Precision farming tackles problems in industry such as the rising global population and their increasing food demands and the pressing need to modernize and streamline farming methods. Typical forms of monitoring in farming are tedious, slow, and unreliable, and often only catch crop diseases in their advanced stages. This work attempts to fill the gaps using swarm drones that work synchronously with each other as a software defined, flying tractor. This work proposes the Embedded Drone Swarm for Precision Agriculture, integrated with Vision AI for surveying crops, disease breakout detection, and outbreaks mitigation. The proposed system has created a Frogeye Leaf Spot, Rust, and Normal (Healthy) detection model for sequential and gradual implementation as a diagnostic system for crop health disease detection aimed at preventative measures. This is the first work that attempts real-time disease detection in both structured and unstructured environments, and addresses the gaps in controlled, efficient and drone-interlinked disease detection with real-time processing of data during or post analysis for informed decisions, all achieved by using inbuilt hardware. The system is built on the MobileNetV2 architecture, using a deep learning model which is integrated along Raspberry Pi, Node MCU, GPS, and a camera, with various components added to the model in post-processing to cloud services. Integration of buffered predictions with confidence scores and voting aims to minimize the false discovery rate and improve the accuracy of the model post processing. The system Automation of disease predictions and recommended treatments is achieved by switching farming apparatus with a relay and HTTP API to be either a water pump or a sprayer. It is further accomplished by sending the IoT Device result to an actuator (or relay) to switch on/off the equipment. For disease prediction and recommended treatments, an automated system was built that alerts a farmer using Twilio services. In conclusion, this system is an intelligent, flexible, and affordable solution for advanced precision agriculture. By providing automated decision-making, real-time crop monitoring, and automated precise intervention for soybean diseases, this solution decreases the need for labor and encourages sustainable agriculture practices.

Precision Agriculture Drone Swarm Vision AI Embedded Systems Plant Disease Detection IoT